1. Field of the Invention
The present invention relates to methods for determining corrected intermodulation distortion (IMD) measurements for a device under test (DUT). The present invention is for use with an appropriate measuring device, such as a vector network analyzer (VNA).
2. Description of the Related Art
Intermodulation distortion (IMD) is defined as nonlinear distortion (i.e., distortion caused by a deviation from a linear relationship between specified input and output parameters of a system or component) characterized by the appearance, in the output of a device, of frequencies that are linear combinations of the fundamental frequencies and all harmonics present in the input signals. It is noted that harmonic components themselves are not usually considered to characterize intermodulation distortion. IMD occurs when the non-linearity of a device with multiple input frequencies causes undesired outputs at other frequencies. In a communications system, for example, this means that signals in one channel can cause interference with adjacent channels.
IMD distortion can be explained with reference to the frequency vs. power graph 100 of FIG. 1. Assume that a signal including two tones at frequencies f1 and f2 is applied to a device under test (DUT). The difference (i.e., offset) in frequency between the two tones can be anywhere between a few kHz and many MHz, although other values are possible. The non-linear characteristics of the DUT generate IMD products, including products at 2f1xe2x88x92f2 and 2f2xe2x88x92f1, which are known as third order products. The relationship of these various spectral components is illustrated in FIG. 1. As shown, the third order IMD products are close to the original tones in frequency, and thus, represent potential adjacent channel spurious signals. Fifth order IMD products occur at 3f1xe2x88x922f2 and 3f2xe2x88x922f1. Seventh order IMD products (not shown) occur at 4f1xe2x88x923f2 and 4f2xe2x88x923f1. Second order IMD products (not shown) occur at f1xe2x88x92f2 and f2xe2x88x92f1. The IMD products of interest are typically the third order products, and possibly the fifth order products. Seventh and second order IMD products are also sometimes of interest. It is recognized that other IMD products, not specifically mentioned here, may also be of interest.
IMD measurements are extremely important in the design and characterization of amplifiers, mixers, passives and other components in communications and other systems. IMD provides a measure of non-linearity and the likelihood that a device will generate signals (due to this non-linearity) that may interfere with other (e.g., adjacent) communications channels. As the density of communications links increase, the requirements grow for lower and lower IMD levels hence making the measurement even more important. At very low IMD levels, the distortion of the measuring receiver itself sometimes limits the measurement. Embodiments of the present invention provides techniques for reducing the effects of those distortions, thus allowing IMD measurements over a wider dynamic range.
Historically, IMD measurements have normally been done in a scalar sense and all measured non-DUT signal products were attempted to be minimized through test set architecture and then neglected. The dynamic range with these setups is often not limited by the noise floor of the receiver, but rather by the IMD products of the receiver or of the source system. While certain techniques have been used to reduce receiver IMD issues, they have limitations.
One possible method for reducing receiver IMD is to pad the receiver input so that signal levels lower, thereby lowering received IMD products. The problem with this solution is that the noise floor is increased.
Another possible method for reducing receiver IMD is to filter the main tones entering the receiver. The problem with this solution is that it restricts measurement to a very narrow frequency range.
Still another possible solution is to feed-forward tone products to cancel IMD at the receiver input. The problem with this solution is that a complex test set is required. Additionally, there are some frequency limitations. Further, the feed-forward must be dynamically adjusted.
Thus, there is a need to correct for receiver IMD products to thereby increase the accuracy of IMD product measurements for DUTs and to allow IMD measurements over a wide dynamic range. The approach for correcting for receiver IMD products preferably overcomes some or all of the above mentioned problems.
The present invention is directed to methods for determining corrected intermodulation distortion (IMD) products for devices under test (DUTs). A ratioed receiver IMD product is measured, where the receiver IMD product results from non-linearities in a receiver of the measuring device (e.g., a vector network analyzer). Next, a ratioed composite IMD product is measured, where the composite IMD product results from non-linearities in both the receiver and the DUT. The corrected DUT IMD product (DUTP) can then be determined by subtracting the ratioed receiver IMD product from the ratioed composite IMD product.
In accordance with an embodiment of the present invention, a first signal having a first frequency (f1) and a second signal having a second frequency (f2) are combined to produce a combined signal having spectral components at the first frequency (f1) and the second frequency (f2). The combined signal is then split (e.g., using a coupler) into a first combined signal and a second combined signal each having spectral components at the first frequency (f1) and the second frequency (f2). The first combined signal is provided to a non-linear device to produce a reference signal having spectral components at the first frequency (f1), the second frequency (f2) and at additional frequencies. The spectral components at the additional frequencies are reference intermodulation distortion (IMD) products. The reference signal is provided to a reference input of the analyzer being used to make the measurements.
A power level of the second combined signal is adjusted to a desired level and provided to a receiver input of the analyzer. At this point, a measurement is made of the second combined signal ratioed to the reference signal, at one of the first frequency (f1) and the second frequency (f2). This measurement is referred to as a ratioed receiver main tone (RM) measurement. A measurement is also made of the second combined signal ratioed to the reference signal, at a frequency associated with an IMD product of interest (e.g., a 3rd order IMD product). This measurement is referred to as a ratioed receiver IMD product (RP) measurement. The RM and RP ratioed measurements effectively calibrate the effects of IMD due to the receiver, enabling the receiver effects to be removed from the DUT measurements.
Now that the system has been calibrated, the second combined signal is provided to the DUT to thereby produce a device output signal. The power level of the second combined signal is adjusted such that the device output signal of the DUT is substantially equal to the desired level. This device output signal is provided to the receiver input of the analyzer. At this point, a measurement can be made of the device output signal ratioed to the reference input signal, at the chosen one of the first frequency (f1) and the second frequency (f2). This measurement is referred to as a ratioed composite main tone (CM) measurement. A measurement is also made of the device output signal ratioed to the reference input signal, at the frequency associated with the IMD product of interest. This measurement is referred to as a ratioed composite IMD product (CP) measurement.
Finally the corrected DUT IMD product (DUTP) can be calculated using the measurements that have been made. In one embodiment, the corrected DUTP is calculated according to the equation: DUTP=CPxe2x88x92RP. In another embodiment, the corrected DUTP is calculated in according with the equation:   DUTP  =      CP    -          RP      ·                        ⅇ                      j            ·                          arg              ⁡                              (                                  CM                  RM                                )                                                    .            
DUTP is the actual DUT intermodulation product ratioed against the reference intermodulation product. However, this may not always be the most useful expression. A somewhat more useful expression would be DUTP ratioed against the DUT output main tone, which should be amplitude corrected. This can be accomplished by also measuring the reference signal power, at the selected one of the first frequency (f1) and the second frequency (f2), to produce an unratioed reference main tone (A1M) measurement. Further, the reference signal power is measured at the frequency associated with the IMD product of interest, to produce an unratioed reference IMD product (A1P) measurement. Then, the DUTP can be ratioed against the composite main tone to eliminate the effects of receiver IMD on signal power measurements.